I have an old 1950s bandsaw (Beaver 2300) that has been refitted, by the previous owner, with a new motor. It runs fine. But now that I am using a power bar with a built in GFCI, the GFCI trips after being on for 15 seconds. When I plug directly into the wall outlet, the 15A breaker does not trip.

Any suggestions on why this should occur, or how I should explore this problem? I have read about ground leakage of electric motors causing nuisance tripping of GFCIs, but I would like to be sure, for safety reasons.

Thanks in advance.

Edit: There is no GFCI on the receptacle.

A shop vac and drill press both plugged into the same GFCI power bar do not trip it. The GFCI seems to be tripping after the bandsaw is on for about 15 seconds, even without cutting. I thought it was only with a heavy load before, but I could be mistaken.

There is no measured resistance between ground and neutral on the motor (ie. open circuit). Likewise for live to ground, and live to neutral with the switch off. When the switch is on (unplugged of course, the resistance is around 1.1 ohms.

I took apart the switch, and verified all the connections. Everything is tight, well connected, and looks quite new. All connections were clean. I vacuumed it all for good measure. Once I put it all back, there was no change.

The motor has no markings on it. I will include a photo. 3 leads go from neutral to the motor, and another 3 go from live to the motor. They are marked 1-6. One of the go to a momentary switch with markings CET36CA-7.

I was told the motor is 1 hp, and it runs off 110v ac. The outlet breaker is 15A and does not trip when the bandsaw is plugged directly into the outlet (ie. no GFCI).

The power bar is Woods Pro Power GFCI 4 outlet https://www.rona.ca/en/electricity-and-lighting/cords-and-timers/power-strips/2. It is UL listed.motor1 motor2 motor junction box motor juntion box 2

Motor connections: motor connections

  • Does the outlet have GFCI? You cannot cascade GFCI units. The downstream GFCI will cause the upstream GFCI to trip, because it actually has a connection between GND and neutral. Note that the breaker for the outlet may be a GFCI type breaker, so even if the outlet appears normal, it may be a GFCI breaker. Take a look at it in the breaker box.
    – mkeith
    Commented Aug 1, 2017 at 3:28
  • No GFCI on the receptacle. But thanks anyhow. A shop vac and drill press both plugged into the same power bar GFCI do not trip it. The bandsaw does, but only when cutting thick hard wood.
    – Fed
    Commented Aug 1, 2017 at 3:32
  • Hmmm. Could it be simple over-current? Or maybe your initial suspicion is correct. Unplug the bandsaw and measure the resistance between the GND prong and neutral prong with an ohmmeter. You could also visually inspect the wiring inside the motor junction box to make sure there is nothing suspicious in there. Like giant wads of electrical tape or oozing goo of some sort. If you have any important updates, add the info to the question, not to a reply here in the comment section.
    – mkeith
    Commented Aug 1, 2017 at 3:36
  • 2
    @mkeith Good point, a neutral-ground fault would tend to only appear when the machine (or circuit) is under load. And frankly when amateurs do wiring, they often treat neutral and ground as the same thing. Commented Aug 1, 2017 at 18:29
  • 1
    I don't suppose you have a clamp meter, do you? It might be worthwhile to split out the separate conductors and measure GND current by itself. You can also put the clamp around both live and hot at the same time. I think this should work. It should just show you the net current (same current that triggers the GFI). You would be looking for a current of 10+ mA, so you would need a sensitive clamp meter. You could do the same thing with the other tools for comparison. If there really is a ground fault, this should flush it out.
    – mkeith
    Commented Aug 2, 2017 at 4:06

4 Answers 4


A 1 Hp on 115/120v line is rated at 16 amps putting a heavy load on the motor could draw 3x the amperage some GFCI's will trip for overload. Shop vacuums rate horse power differently than true HP rated motors. A megger is needed to check for winding leakage. Most ohm meters are very low voltage where a megger is a high voltage ohm meter. With a megger you connect 1 lead to the case and 1 to the winding I usually use the 500v setting when checking 120v motors.

  • I bought an inexpensive megaohmmeter ($70). Could you please let me know how to test for winding leakage? I updated the image of the motor connections.
    – Fed
    Commented Aug 21, 2017 at 21:22
  • Connect 1 lead to the case and 1 to the hot lead to the motor. With home wiring the 500v setting would be more than enough press test , charge or crank and see what the resistance value is.
    – Ed Beal
    Commented Aug 21, 2017 at 23:45
  • So, basically ac hot and ac ground (since the case it connected to ground)?
    – Fed
    Commented Aug 21, 2017 at 23:50
  • Yes if there is a low value that identifies there can be leakage the next step is finding where. the motor windings, capacitor, wiring, switch are all possible locations.
    – Ed Beal
    Commented Aug 21, 2017 at 23:56
  • Here are the results: 100V >110M ohms, 250V >5.5G ohms, 500V >5.5G ohms, 1000V 4.5G ohms (but slowly rising to 5.4G ohms. Looks good?
    – Fed
    Commented Aug 22, 2017 at 0:16
> never gave me any trouble until I added ground fault detection 
> everything else works
> this one fails
> only possibility 
> ground fault detector must be broken

This is what everyone says every single time they get a GFCI trip. Seriously. People just can't believe their stuff has ground faults. It does, and that's why we have ground fault protection.

Yes, you changed the motor. What about the switch, interlocks or other controls? Could they be dusty, oily, full of sawdust, in need of a cleanout?

Is the motor a quality machine (e.g. Made in USA)? Speaking of that, is the power strip a quality machine?

A GFCI power strip isn't really Code anyway. A garage requires GFCI protection at the receptacle or breaker. These tend to be better units as they are made by mid-tier receptacle makers (e.g. Leviton) or top-tier panel makers (e.g. GE, Siemens).

  • I did not say it is an issue with the GFCI. I think it is working fine. I am trying to find the problem. Also, I did not change the motor, the original owner did. It is not in the garage.
    – Fed
    Commented Aug 1, 2017 at 17:15
  • Oh, the GFCI requirement applies to most utility areas in a home. The "other guy did it" certainly suggests it's worth a very thorough inspection of his work. Commented Aug 1, 2017 at 18:32

Not all GFCIs are created equal. Cheaper GFCIs are susceptible to being "fooled" by capacitance in a circuit, and your AC induction motor looks to be a Capacitor Start at least, probably Cap-Start / Cap-Run, meaning there are multiple capacitors in it and the Run cap is always in the circuit. it takes a while for a capacitive charge to build up across the rotor and stator to where it discharges to ground, which is what the GFCI sees and reacts to. The more sophisticated GFCIs have filtering to discriminate those sorts of events and not over react to them, cheaper ones don't.

Your vacuum cleaner likely has what is called a "universal" motor, i.e. the kind with brushes, so no capacitors. The drill press may be Split Phase, maybe even a universal motor too, or just Cap-Start / Induction Run, so the capacitor is only in the circuit for a second or two, it's hard to say.

I wouldn't run a table saw from a GFCI power strip anyway, there is no need.


A shop vac and drill press both plugged into the same power bar GFCI do not trip it. The bandsaw does, but only when cutting thick hard wood.

It sounds like the heavy load of the bandsaw motor is shifting the current out of phase enough to trip the GFCI. GFCI's can sometimes have problems with inductive loads like motors and lighting ballasts.

Circuit breakers and GFCI on power strips are not as well manufactured as receptacles and circuit breakers. Their tolerances are probably really bad. Personally, I would not trust them to protect me or my family. You need a higher quality GFCI.

If you have a shop area that is unfinished or in a garage, it should all be protected by GFCI. Either use the load side of a GFCI receptacle to protect other receptacles or install a GFCI breaker for all circuits.

Once you have done this, if the bandsaw still trips, I would say the GFCI it has a problem and it should be repaired or replaced. Since it is only tripping during heavy loading I would say it is the power strip that is suspect not the bandsaw.

After you posted the 1 HP spec of the motor I would suspect the power strip's circuit breaker is actually what's tripping. Not the GFCI.

Good luck and stay safe!

  • I don't see why a GFCI would trip due to inductive load. The hot and neutral currents are not out of phase with each other due to inductive load. They are always exactly 180 degrees out of phase, no matter the load.
    – mkeith
    Commented Aug 1, 2017 at 16:43
  • You need to re-study your theory. A purely inductive load shifts the current 90°out of phase. So an inductive load always creates some phase shift depending on the load. It may be more likely that the power strip circuit breaker is tripping from excess current drawn by the saw.
    – ArchonOSX
    Commented Aug 2, 2017 at 0:34
  • 1
    The phase shift you are talking about is between current and voltage. That is not what trips a GFCI. What trips a GFCI is a difference between neutral and hot wire currents. Unless there is a ground fault, those two currents are always exactly 180 degrees out of phase and equal in magnitude, no matter what the load impedance is. Resistive, capacitive or inductive. This is just basic Kirchoff's Current Law, which derives from conservation of charge.
    – mkeith
    Commented Aug 2, 2017 at 3:24
  • True, that's not what is supposed to trip a GFCI. The phase shift can fool a poorly made GFCI. A well made one is not so fooled.
    – ArchonOSX
    Commented Aug 2, 2017 at 10:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.